A variety of evidence suggests serotonin 1A (5-HT1A) receptor function is abnormal in panic disorder (PD), postraumatic stress disorder (PTSD), and depression. This protocol investigates the pharmacological basis for 5-HT1A abnormalities in vivo by applying positron emission tomography (PET) and the 5-HT1A receptor radioligand [F-18]FC-WAY100635 to assess 5-HT1A binding potential in panic disorder, postraumatic stress disorder, and unipolar and bipolar depression. Because central 5-HT1A receptor density is down-regulated in rodents by corticosterone administration and by stress-mediated corticosterone secretion, assessments of HPA-axis activity will be assessed to determine whether down-regulation of 5-HT1A receptors correlates with cortisol hypersecretion in PD, PTSD, and MDD. PET images of 5-HT1A binding are acquired in subjects who are unmedicated PD, PTSD and depressed patients, or are healthy volunteers. The 5-HT1A receptor volume of distribution (V) is calculated with a two tissue compartment model of PET data after a bolus infusion of ([F-18]FCWAY). The 5-HT1A receptor V in selected regions of interest (ROI) are the hippocampus, amygdala, ventral ACC, and orbital cortex, and are compared between the entire depressive and control samples. The relationships between 5-HT1A receptor V and salivary cortisol, and between 5-HT1A receptor V and cerebrospinal fluid concentrations of corticotrophin releasing hormone (CRH), are assessed by linear regression analysis. During the past year, 52 subjects have been entered into this study. We have developed satisfactory compartmental models for pixelwise determination of V and have developed a method for defining regions of interest that will be semi-automated and thus independent of operator bias. Sample sizes in each group remain too small to perform statistical analyses on the imaging data. However, neuropsychological assessments of the study?s depressed and healthy subjects have shown deficits in the unmedicated unipolar and bipolar depressed patients that are consistent with dysfunction of the amygdala, ventral anterior cingulate cortex, and orbital cortex. Specifically, unmedicated depressed patients displayed deficits in inhibiting inappropriate responses to affective stimuli that was not present in the healthy control sample suggesting an attentional bias towards these stimuli. The unmedicated depressed sample also required more time to deliberate on a risk-based task, consistent with findings in frontal lesion patients.